verify_cad_queryLABSCO SUMMARY
The part is code, so the review can be automated.
Conversational CAD produces plausible Python that quietly makes the wrong shape. Running the file and checking it against the description you started from turns that into a verdict rather than a surprise at the printer.
What it is
A verification step for parametric CAD written in CadQuery: point it at a Python model file and a description of what the part should be, and get a verdict back with the reasoning.
What you get
- A CadQuery model file validated against criteria you write in plain language
- A PASS or FAIL result carrying a message, the file checked, the criteria used, and further detail
- CadQuery underneath, with STL and STEP export for 3D printing and CAD handoff
- SVG rendering for looking at the model
Requirements
uv, with the cad extra installed — plain uv sync leaves the verification path without its dependencies. A second tool for generating CadQuery scripts from a description is present but returns a stub, so write the model yourself and use this to check it.
Setup effort
Build from source
USE CASES
01
Confirming a generated model matches what you asked for
Run the file against the description — a mug with a handle, 10cm tall, 8cm across — and get a pass or fail.
02
Catching a bad edit before printing
Re-check the file after a dimension change rather than discovering it in the slicer.
03
Iterating on a shape in conversation
Adjust the script, verify, read the failure detail, adjust again.
04
Handing a part to someone else
Verify against the written criteria first, so the file you send has been checked against its own brief.
2 tools
Listed in project docs. Your client may show a different list.
generate_cad_queryBEFORE YOU INSTALL
What to know first
Tools you get
2 tools named in the project's documentation.
Requires
- uv
- Python 3.11 or later
Where it runs
The documented client runs the server on your machine.
INSTALLATION
Connect your client
MCP Clients
Claude Desktop
CLAUDE DESKTOP · JSON
{
"mcpServers": {
"cadquery-server": {
"command": "python",
"args": [
"<YOUR_PATH_TO_CADQUERY_MCP_SERVER>/server.py"
],
"env": {}
}
}
}
